seedance-2-0
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a detailed manual for generating cinematic video content. It includes prompt templates, camera control instructions, and parameter guidance for third-party AI video models.
- [DATA_EXPOSURE]: The skill correctly identifies the need for environment variables (e.g., FAL_KEY, REPLICATE_API_TOKEN) to manage API credentials. It does not contain hardcoded secrets, nor does it include commands that would exfiltrate these keys.
- [PROMPT_INJECTION]: While the skill uses strong instructional language such as 'CRITICAL' and 'IMPORTANT', these are used exclusively to guide the user in crafting effective prompts for the video generation model and do not attempt to override the AI agent's core safety guidelines or system instructions.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface where user-provided prompts are passed to video generation tools. While this constitutes a data ingestion point, the skill itself does not provide an exploitable implementation that would lead to unauthorized code execution on the host environment.
- [EXTERNAL_DOWNLOADS]: The skill references several well-known and reputable AI service providers and documentation sources (e.g., fal.ai, Replicate, ByteDance, Wikipedia). These references are informational and do not involve the silent downloading of executable scripts.
Audit Metadata